Merge remote-tracking branch 'origin/develop' into aosewski/ggemm_multi_d2

cfc2be07 · Adam Osewski · 30e4f4eb · 497ccb87 · cfc2be07 · cfc2be07
Commit cfc2be07 authored Jul 03, 2024 by Adam Osewski
17 changed files
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_bias_fastgelu_bf16_i8_bf16_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_bias_fastgelu_bf16_i8_bf16_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_fastgelu_bf16_i8_bf16_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_fastgelu_bf16_i8_bf16_mk_kn_mn_instance.cpp
--- a/profiler/include/profiler/profile_grouped_gemm_multiply_tile_loop_impl.hpp
+++ b/profiler/include/profiler/profile_grouped_gemm_multiply_tile_loop_impl.hpp
--- a/profiler/src/CMakeLists.txt
+++ b/profiler/src/CMakeLists.txt
--- a/profiler/src/profile_grouped_gemm.cpp
+++ b/profiler/src/profile_grouped_gemm.cpp
--- a/profiler/src/profile_grouped_gemm_multiply_tile_loop.cpp
+++ b/profiler/src/profile_grouped_gemm_multiply_tile_loop.cpp
--- a/test/CMakeLists.txt
+++ b/test/CMakeLists.txt
--- a/test/contraction/test_contraction_xdl.cpp
+++ b/test/contraction/test_contraction_xdl.cpp
--- a/test/grouped_convnd_bwd_weight/test_grouped_convnd_bwd_weight.cpp
+++ b/test/grouped_convnd_bwd_weight/test_grouped_convnd_bwd_weight.cpp
--- a/test/grouped_convnd_fwd/CMakeLists.txt
+++ b/test/grouped_convnd_fwd/CMakeLists.txt
--- a/test/grouped_convnd_fwd/test_grouped_convnd_fwd_xdl_wmma.cpp
+++ b/test/grouped_convnd_fwd/test_grouped_convnd_fwd_xdl_wmma.cpp
--- a/test/grouped_convnd_fwd/test_grouped_convnd_fwd_multi_ab_interface.cpp
+++ b/test/grouped_convnd_fwd/test_grouped_convnd_fwd_multi_ab_interface.cpp
--- a/test/smfmac_op/CMakeLists.txt
+++ b/test/smfmac_op/CMakeLists.txt
--- a/test/smfmac_op/smfmac_op.cpp
+++ b/test/smfmac_op/smfmac_op.cpp
--- a/test/smfmac_op/smfmac_op_util.hpp
+++ b/test/smfmac_op/smfmac_op_util.hpp
--- a/test/smfmac_op/smfmac_op_xdl.cpp
+++ b/test/smfmac_op/smfmac_op_xdl.cpp
--- a/test/wmma_op/wmma_op_util.hpp
+++ b/test/wmma_op/wmma_op_util.hpp